Start free

Google Meet transcription.From Drive to speaker-labeled notes.

Drop the MP4 Google Meet dropped into your Drive. Get a speaker-labeled, timestamped transcript with an AI summary — no Workspace upgrade, no Gemini add-on, works on any Google Meet recording.

Drop a file, or pick one

MP3 · WAV · M4A · MP4 · MOV · MKV · OGG · OPUS · FLAC · WEBM — up to 100 MB anonymously

Paste a link, we’ll fetch the audio

YouTube · TikTok · Vimeo · Twitter · SoundCloud · Spotify · 50+ more

Record straight from your browser

Sign up takes 30 seconds — recording opens right after, in the dashboard.

No card required~90s per 60-min fileSRT · VTT · DOCX · TXTFiles auto-deleted in 24h

↓ Watch what comes out

Drive MP4 in. Clean transcript out.

Google Meet always records to a single mixed-down MP4 in the organiser's Drive — no per-participant track option exists. We run acoustic diarization on that mono file and split speakers from voice alone.

Meet recording from DriveREC 4 speakers · 32:14
auto-detected en-GB48 kHz mono · AAC 192 kbps
~90s
Transcript · streaming93% accuracy
S1

Before we wrap — Priya, did onboarding sign off on the new welcome flow?

S2

They did, with one change. They want the verification email moved to step two.

S3

That's fine on our side. I can ship it Friday if QA clears Thursday.

S1

Great. Let's lock Friday and I'll tell support to expect the change.

93% on mono Drive MP4DOCX · SRT · TXT · JSON

↓ This is the dashboard

This is what loads when the job finishes.

Same layout as the real dashboard — Summary, full Transcript, Speakers tab, Exports. Key points and action items extracted automatically. Auto-tags on every job.

Try it on your own file — it's free

Three real options · honest comparison

Google's built-in. Tactiq or Fireflies. Or us.

Google ships transcripts and Gemini notes on higher Workspace tiers. Tactiq and Fireflies live as a Chrome extension or calendar bot. We work with the MP4 already sitting in your Drive — no upgrade, no extension, no calendar access.

Option 01

Google Meet built-in

Live transcripts and Gemini 'take notes' inside Meet. Gated behind specific Workspace SKUs.

RequiresWorkspace Business Standard+
Speaker diarizationYes, on live transcript
Languages~15 for live captions
ExportGoogle Doc only
AI summaryGemini add-on or AI tier
Cost$14+/user/mo Workspace
Best forTeams already on a paid Workspace tier that meet exclusively in English-ish and live inside Google Docs.
Option 02

Transcription.Solutions

Drop the Drive MP4. Or send a bot to the meeting URL. Works on any Workspace plan — including free Gmail accounts.

RequiresNothing on Google side
Speaker diarizationAcoustic, 2-10 speakers
Languages99, auto-detected
ExportDOCX · SRT · VTT · TXT · JSON
AI summaryFree on every plan
Cost · per min$0.03
Best forAnyone on a non-Enterprise Workspace plan, multilingual teams, or anyone who wants the transcript out of Google Docs as DOCX or SRT.
Option 03

Tactiq / Fireflies

Chrome extension that grabs Meet captions, or a calendar bot that joins the call.

RequiresExtension or calendar OAuth
Speaker diarizationCaption-based, EN-tuned
LanguagesInherits Meet caption set
ExportFree tier limits transcripts/mo
AI summaryBehind paid tier
Cost$10–19/user/mo
Best forEnglish-only teams that prefer a live-captioning Chrome extension over working with the saved recording.

Pricing and feature flags accurate as of May 2026. Google Meet built-in transcripts depend on Workspace edition and admin policy.

Specific to Google Meet

Three things that bite people on generic transcription tools.

Meet's recording behaviour is fixed — but a few job-form flags clean up most of the mess.

What goes wrong

  1. 1Mono mix-down is forced. Meet never separates participants into channels — every generic tool that assumes stereo split (most do) falls back to weak acoustic guesses.
  2. 2The Drive MP4 has a long stretch of silence at the start while the recording bot finishes joining. Generic tools count it as one giant turn for 'Speaker 1'.
  3. 3Internal project names (Northstar, Helios, Beacon) get spelled phonetically — and Gemini's own summary inherits those misspellings.

What to flip here

  1. 1Pick Conversational · 2-10 speakers on the job form. Our diarizer is tuned for mono mixed-down meeting audio, not stereo podcast files.
  2. 2Leave Trim leading silence on (default). We cut any silence longer than 5 seconds before diarization starts, so the first real speaker is Speaker 1.
  3. 3Paste project names into Custom vocabulary. The recognizer treats them as hints, and the AI summary uses the corrected spellings.

Recommended job settings for Google Meet

Drop a Meet MP4 from Drive and these flip on by default. Override per-job from the form.

Diarization
Acoustic · mono-tuned
Speaker model
Conversational · 2-10 speakers
Language
Auto-detect · multi-lingual on
Leading silence
Trim > 5s
Summary
Decisions + action items
Export
DOCX · SRT · timestamped TXT

Accuracy · real-world numbers

93% on a typical Drive recording. Higher with clean mic audio.

Google Meet writes a single mono MP4 to Drive — there's no per-channel option to lean on, so the floor and ceiling are set by mic discipline and how distinct the voices are. Numbers below are from real customer Meet files, not lab benchmarks.

95%+
≤3 speakers, headset mics

Small standups or 1:1s where everyone wears a headset. Distinct voice profiles, near-zero background noise.

93%
4-6 speakers, laptop mics

Typical Meet call. Acoustic diarization holds; expect a 1-2 minute pass to rename speaker chips at the start.

89%
7+ speakers or one in a meeting room

Conference-room mic picks up multiple people on one channel. Diarization may merge the room speakers — overlap drops accuracy.

85%
Phone dial-in or weak Wi-Fi participant

Meet downsamples that leg. Words usable, occasional misses on numbers, acronyms, and proper nouns. Worst case in our data.

Common questions

8 things people ask about Google Meet transcription.

01Can you pull from a Google Drive link directly?+
Yes, if you share the file as 'Anyone with the link → Viewer'. Otherwise download the MP4 from Drive and drop it on the upload page — takes about 20 seconds. We don't ask for Drive OAuth.
02Do I need Workspace Business Standard for this to work?+
No. Meet records to Drive on every paid Workspace tier (and on free personal accounts with the host's Drive). We work with whatever MP4 lands there — the built-in live transcript feature is a separate thing we don't depend on.
03How does this compare to Gemini's 'Take notes for me'?+
Gemini summarises live and writes to a Google Doc. We work from the saved recording, so you get a verbatim transcript with timestamps plus a summary, exportable as DOCX or SRT. And we don't need an AI add-on SKU.
04Why does my transcript say 'Speaker 1' for two different people?+
Meet's mono mix-down makes acoustic diarization the only option, and two people on similar laptop mics in similar rooms can blur. Rename the chip once in the editor and the change propagates through the whole transcript and the summary.
05Can a bot join my Meet call live instead?+
Yes. Paste the meet.google.com URL into the dashboard and a bot joins as a participant. The host has to admit it from the lobby — Meet doesn't allow silent joins. The transcript streams while the call runs.
06What about the captions/transcript file Meet already saves?+
Meet's own transcript (when enabled) lands in Drive as a Google Doc with speaker turns but no second-level timestamps or SRT export. If you already have it, you can upload it as a reference — but transcribing the MP4 directly is usually cleaner.
07What languages work?+
99, auto-detected from the audio. Meet's own live captions cover roughly 15 — if your team meets in Portuguese, Polish, or Turkish, the built-in transcript will struggle where ours won't.
08Where does the recording go after I upload it?+
The MP4 is deleted from our servers within 24 hours of the job completing. The transcript and summary stay in your account until you delete them. We don't train models on customer audio.

Drop your Meet recording. See what comes out.

30 free minutes every month. No card. Speaker labels, 99 languages, AI summary, all exports included.

Start free